A community for electronic music production, modular synthesis, sdiy and more. Log inSign up with Discord, Facebook, Google, Twitter
jp avatar

Vimeo
i trained a neural network to recognize my voice speaking the german vowels a, e, i, o, u. each time it recognizes a vowel it drops the vowel as a letter and displays the confidence of the recognition.

1 3 55 Jul 20 Vimeo

Comment

tylth avatar

That's confidence down to the fourteenth digit! I have that too : )

This would be a fun party game!

2 Jul 21
jp avatar
jp

@tylth yeah it's really very confident. :-)
i haven't tested it with other people's voices yet but it also works quite well with the internal mic of my mac instead of the sm58 so it's a little bit robust.

Jul 23
jp avatar
jp

@tylth and yes it's actually planned to use that in an interactive installation. for arts exhibitions, parties, etc.

Jul 23
What do you think? Sign up with Discord, Facebook, Google, Twitter to leave a comment.